arXiv subjects

Binglei Lou

Publications and source records attributed to Binglei Lou.

9 recordsLinked to original sources

arXiv2026

MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs

Emerging agentic LLM workloads are driving rapidly growing demand on both memory capacity and bandwidth, with different phases of inference (e.g., prefill and decode) imposing distinct requirements. Industry is responding by composing heterogeneous accelerators into single interconnected systems, as exemplified by NVIDIA's Vera Rubin platform, where each device brings its own memory architecture. This heterogeneity is further compounded by a widening landscape of available memory technologies: high-density on-chip SRAM, HBM, LPDDR, GDDR, and emerging options such as high-bandwidth flash (HBF), each offering different capacity, bandwidth, and power trade-offs. Identifying the right memory architecture for next-generation inference accelerators requires navigating a vast and rapidly evolving design space, in which the interplay between workload characteristics, NPU design dimensions, and memory system design remains largely underexplored. To address this challenge, we present MemExplorer, a new memory system synthesizer for heterogeneous NPU systems. MemExplorer provides a unified abstraction for modeling diverse memory technologies across different hierarchy levels (e.g., on-chip and off-chip) and automatically determines an efficient heterogeneous memory system together with NPU design choices (e.g., matrix engine size) to balance throughput and power between prefilling and decoding devices in a multi-device NPU system. Experimental results show that, under the same power budget for agentic workloads, MemExplorer achieves up to 2.3x higher energy efficiency than the baseline NPU and 3.23x higher than H100 in the prefill-only setting. Under equivalent performance targets in the decode setting, it further delivers up to 1.93x and 2.72x higher power efficiency over the baseline NPU and H100, respectively.

cs.AR↗

arXiv2026

KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware

New AI accelerators with novel instruction set architectures (ISAs) often require developers to manually craft low-level kernels, a time-consuming and error-prone process that does not scale across hardware targets. This delays emerging hardware platforms from reaching the market. While prior LLM-based code generation has shown promise in mature GPU ecosystems, it remains unclear whether agentic LLM systems can quickly produce valid and efficient kernels for emerging hardware with new ISAs. We present KernelCraft: the first benchmark for evaluating an LLM agent's ability to generate and optimize low-level kernels for customized accelerators through a function-calling, feedback-driven workflow. We evaluate agent performance across three emerging accelerators on more than 20 machine-learning tasks, each with five diverse task configurations. Across four leading reasoning models, the strongest agents generate functionally correct kernels for unseen ISAs within a few refinement steps and produce optimized kernels that match or outperform compiler baselines. These results demonstrate KernelCraft's potential to accelerate the accelerator chip development cycle. KernelCraft is available at https://kernelcraft-cam.github.io/.

cs.AR↗

arXiv2026

NPU Design for Diffusion Language Model Inference

Diffusion-based LLMs (dLLMs) fundamentally depart from traditional autoregressive (AR) LLM inference: they leverage bidirectional attention, block-wise KV cache refreshing, cross-step reuse, and a non-GEMM-centric sampling phase. These characteristics make current dLLMs incompatible with most existing NPUs, as their inference patterns, in particular the reduction-heavy, top-$k$-driven sampling stage, demand new ISA and memory hierarchy support beyond that of AR accelerators. In addition, the blocked diffusion KV cache breaks from the append-only paradigm assumed by AR NPUs, and conventional AR-derived KV quantization schemes were designed for static activation distributions and do not account for the step-wise distribution shifts introduced by iterative block-wise refinement in dLLMs. In this paper, we introduce the first NPU accelerator specifically designed for dLLMs. It delivers: a dLLM-oriented ISA and compiler; a hardware-optimized execution model for both the transformer inference and diffusion sampling used in dLLMs; a novel Block-Adaptive Online Smoothing (BAOS) for quantizing KV cache in dLLMs; and a complete RTL implementation synthesized in 7nm. To evaluate and validate our design, we introduce a tri-path simulation framework that comprises analytical, cycle-accurate, and accuracy simulators, together with cross-validations against physical hardware. The full NPU stack, including ISA, simulation tools, and quantization software, will be open-sourced upon acceptance.

cs.AR↗

arXiv2026

Enhancing LUT-based Deep Neural Networks Inference through Architecture and Connectivity Optimization

Deploying deep neural networks (DNNs) on resource-constrained edge devices such as FPGAs requires a careful balance among latency, power, and hardware resource usage, while maintaining high accuracy. Existing Lookup Table (LUT)-based DNNs -- such as LogicNets, PolyLUT, and NeuraLUT -- face two critical challenges: the exponential growth of LUT size and inefficient random sparse connectivity. This paper presents SparseLUT, a comprehensive framework that addresses these challenges through two orthogonal optimizations. First, we propose an architectural enhancement that aggregates multiple PolyLUT sub-neurons via an adder, significantly reducing LUT consumption by 2.0x-13.9x and lowering inference latency by 1.2x-1.6x, all while maintaining comparable accuracy. Building upon this foundation, we further introduce a non-greedy training algorithm that optimizes neuron connectivity by selectively pruning less significant inputs and strategically regrowing more effective ones. This training optimization, which incurs no additional area and latency overhead, delivers consistent accuracy improvements across benchmarks -- achieving up to a 2.13% gain on MNIST and 0.94% on Jet Substructure Classification compared to existing LUT-DNN approaches.

cs.AR↗

arXiv2025

Low-Latency FPGA Control System for Real-Time Neural Network Processing in CCD-Based Trapped-Ion Qubit Measurement

Accurate and low-latency qubit state measurement is critical for trapped-ion quantum computing. While deep neural networks (DNNs) have been integrated to enhance detection fidelity, their latency performance on specific hardware platforms remains underexplored. This work benchmarks the latency of DNN-based qubit detection on field-programmable gate arrays (FPGAs) and graphics processing units (GPUs). The FPGA solution directly interfaces an electron-multiplying charge-coupled device (EMCCD) with the subsequent data processing logic, eliminating buffering and interface overheads. As a baseline, the GPU-based system employs a high-speed PCIe image grabber for image input and I/O card for state output. We deploy Multilayer Perceptron (MLP) and Vision Transformer (ViT) models on hardware to evaluate measurement performance. Compared to conventional thresholding, DNNs reduce the mean measurement fidelity (MMF) error by factors of 1.8-2.5x (one-qubit case) and 4.2-7.6x (three-qubit case). FPGA-based MLP and ViT achieve nanosecond- and microsecond-scale inference latencies, while the complete single-shot measurement process achieves over 100x speedup compared to the GPU implementation. Additionally, clock-cycle-level signal analysis reveals inefficiencies in EMCCD data transmission via Cameralink, suggesting that optimizing this interface could further leverage the advantages of ultra-low-latency DNN inference, guiding the development of next-generation qubit detection systems.

quant-ph↗

arXiv2025

Combating the Memory Walls: Optimization Pathways for Long-Context Agentic LLM Inference

LLMs now form the backbone of AI agents across a diverse range of applications, including tool use, command-line interfaces, and web or computer interaction. These agentic LLM inference tasks are fundamentally different from chatbot-focused inference. They often involve much longer context lengths to capture complex and prolonged inputs, such as an entire webpage DOM or complicated tool-call trajectories. This, in turn, generates significant off-chip memory traffic during inference and causes workloads to be constrained by two memory walls, namely the bandwidth wall and the capacity wall, preventing compute units from achieving high utilization. In this paper, we introduce PLENA, a hardware-software co-designed system built around three core optimization pathways. PLENA features a novel flattened systolic-array architecture (Pathway 1) and efficient compute and memory units that support an asymmetric quantization scheme (Pathway 2). It also provides native support for FlashAttention (Pathway 3). In addition, PLENA includes a complete software-hardware stack, consisting of a custom ISA, a compiler, a transaction-level simulator, and an automated design-space exploration flow. Experimental results show that PLENA delivers up to 2.23x and 4.70x higher throughput than the A100 GPU and TPU v6e, respectively, under identical multiplier counts and memory configurations during LLaMA agentic inference. PLENA also achieves up to 4.04x higher energy efficiency than the A100 GPU. The full PLENA system, including its simulator, compiler, ISA, and RTL implementation, will be open-sourced to the research community.

cs.AR↗

arXiv2025

SparseLUT: Sparse Connectivity Optimization for Lookup Table-based Deep Neural Networks

The deployment of deep neural networks (DNNs) on resource-constrained edge devices such as field-programmable gate arrays (FPGAs) requires a careful balance of latency, power, and resource usage while maintaining high accuracy. Existing Lookup Table (LUT)-based DNNs, including LogicNets, PolyLUT, PolyLUT-Add, and NeuraLUT, exploit native FPGA resources with random sparse connectivity. This paper introduces SparseLUT, a connectivity-centric training technique tailored for LUT-based DNNs. SparseLUT leverages a non-greedy training strategy that prioritizes the pruning of less significant connections and strategically regrows alternative ones, resulting in efficient convergence to the target sparsity. Experimental results show consistent accuracy improvements across benchmarks, including up to a 2.13\% increase on MNIST and a 0.94\% improvement for Jet Substructure Classification compared to random sparsity. This is done without any hardware overhead and achieves state-of-the-art results for LUT-based DNNs.

cs.AR↗

arXiv2024

PolyLUT-Add: FPGA-based LUT Inference with Wide Inputs

FPGAs have distinct advantages as a technology for deploying deep neural networks (DNNs) at the edge. Lookup Table (LUT) based networks, where neurons are directly modeled using LUTs, help maximize this promise of offering ultra-low latency and high area efficiency on FPGAs. Unfortunately, LUT resource usage scales exponentially with the number of inputs to the LUT, restricting PolyLUT to small LUT sizes. This work introduces PolyLUT-Add, a technique that enhances neuron connectivity by combining $A$ PolyLUT sub-neurons via addition to improve accuracy. Moreover, we describe a novel architecture to improve its scalability. We evaluated our implementation over the MNIST, Jet Substructure classification, and Network Intrusion Detection benchmark and found that for similar accuracy, PolyLUT-Add achieves a LUT reduction of $2.0-13.9\times$ with a $1.2-1.6\times$ decrease in latency.

cs.LG↗

arXiv2024

fSEAD: a Composable FPGA-based Streaming Ensemble Anomaly Detection Library

Machine learning ensembles combine multiple base models to produce a more accurate output. They can be applied to a range of machine learning problems, including anomaly detection. In this paper, we investigate how to maximize the composability and scalability of an FPGA-based streaming ensemble anomaly detector (fSEAD). To achieve this, we propose a flexible computing architecture consisting of multiple partially reconfigurable regions, pblocks, which each implement anomaly detectors. Our proof-of-concept design supports three state-of-the-art anomaly detection algorithms: Loda, RS-Hash and xStream. Each algorithm is scalable, meaning multiple instances can be placed within a pblock to improve performance. Moreover, fSEAD is implemented using High-level synthesis (HLS), meaning further custom anomaly detectors can be supported. Pblocks are interconnected via an AXI-switch, enabling them to be composed in an arbitrary fashion before combining and merging results at run-time to create an ensemble that maximizes the use of FPGA resources and accuracy. Through utilizing reconfigurable Dynamic Function eXchange (DFX), the detector can be modified at run-time to adapt to changing environmental conditions. We compare fSEAD to an equivalent central processing unit (CPU) implementation using four standard datasets, with speed-ups ranging from $3\times$ to $8\times$.

cs.AR↗